Research On Microgrid Optimization Based On Improved NSGA2 Algorithm

With the rapid development of society and the increasing awareness of environmental protection,the contradiction between energy and environmental protection is becoming clearer than ever.The over-exploitation of fossil energy has damaged the natural environment.In order to settle the contradiction of environmental protection and the increasing per capita resource consumption,the exploitation and the utilization of renewable energy such as wind energy and solar energy is already underway.Clean energy,with the advantages of large development reserves and environmentally frendly,but is greatly affected by the climate and its output is random.Therefore,the microgrid,which regarded as a platform that consumes clean energy come at the right time.As a small power generation and distribution system,the microgrid can not only solve new energy consumption problem,but be able to switch between grid-connected and isolated grid modes flexinly when the system fails.At present,the microgrid is still in the early stage of development,and there are exist problems of coordination control difficulty and high comprehensive cost.The optimized operation of the microgrid has made an important difference in improving the utilization of clean energy,reducing the comprehensive cost,and rationally configuring the distributed power supply.Therefore,it is of great significance to study the optimal operation of the microgrid.First,according to the second-generation Non-dominated Sorting Genetic Algorithm(NSGA2),with the poor convergence accuracy and the limitation of the high calculation complexity,puts forward GNSGA2 algorithm(G Non-dominated Sorting Genetic Algorithm 2,GNSGA2).The Algorithm in the initial population mixing method,optimization strategy,make the population distribution uniformity and individual selection method with the directional differential mechanism in the process of genetic evolution,which improves the global search ability and thus increases the microgrid.The number of solutions after the model is solved.The two algorithms are applied to the comparison of ZDT and DTLZ series functions tests.The simulation results verify the effectiveness of the improved algorithm.Secondly,it introduces the current three typical structures and working principles of the microgrid,and mainly elaborates the basic principles and mathematical models of the distributed power sources in the microgrid system,including solar panels,wind turbines,micro-turbines and energy storage devices,which laid a theoretical foundation for the establishment of a multi-objective optimal operation model of microgrid.Finally,take the minimum comprehensive cost of the microgrid as the objective function one;the minimum exchange power of the large power grid and the minimum 24-hour load shedding are taken as the objective function two of the grid-connected model and the isolated grid model.It proposes the correlation equation and inequality constraints,establishes the multi-objective optimal operation model of microgrid.According to the respective operating characteristics of the grid-connected and isolated network modes,corresponding operating strategies are developed.The GNSGA2 algorithm is applied to optimization model to obtain the Pareto solution set of the two objective functions the iteration diagram under the single objective function and the optimal output force of each distributed power supply in 24 hours,and compare the results with those obtained by the NSGA2 algorithm.Simulation res`ults show that the improved GNSGA2 algorithm increases the number and range of decentralized solutions in the set,and speed up the convergence comparatively.